JP5217960B2 - Image processing apparatus, image processing method, and program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program.

  There are cases where an operator desires to extract a region desired by an operator, such as a character, a photograph, a figure, or a table (referred to as an image element or an image component) in a document image (copy operation or the like). Thus, when the range (area) is specified so that the operator surrounds it with a mouse or the like, it can be said that it is generally difficult to accurately specify and extract the four corners of the image element. Therefore, in practice, it is often the case that an area to be roughly extracted is designated by taking an area larger or smaller than a desired range.

  Regarding the clipping of the designated area, for example, it is ideal that the designation operation can be performed so as to cut out only the necessary character area from the document image without including the useless area as shown in FIG. In many cases, an extra portion (area) is included as shown in FIG. 2 or FIG. 3, or an insufficient portion appears on the contrary.

By the way, there is a technology of an image processing apparatus that extracts and processes a region from a document image. For example, Patent Document 1 describes a technique of a rectangle extraction method that cuts out a character string by generating a rectangle surrounding connected runs that are a group of line elements. Further, Patent Document 2 describes a technique such as an area designation method for extracting a document area including an instructed point based on a character interval or the like in a document image. Patent Document 3 discloses an invention of a region extraction method for a document image in which region extraction processing is performed using an integrated threshold set based on a white run histogram that is a series of white pixels in a binarized image. .
JP-A-3-144873 JP-A-3-204781 JP-A-5-40849

  However, in the inventions described in Patent Documents 1 to 3, when an operator designates a range to be selected by a device such as a mouse or a touch pen, an area to be selected by the operator is determined based on the designation. Acquisition is not considered.

  In other words, when the character area to be selected is a rectangle, it is difficult for an operator to specify such a range accurately by using an input device. May be a large or small rectangle or a closed multi-curve (closed figure formed by connecting a plurality of curves), for example, as shown in FIG. There was a problem that an extra area or an insufficient area appeared.

  In addition, when the region extraction process is performed using the integrated threshold set based on the histogram, if a line such as a ruled line is included in the designated region, it is dragged by this and the region cannot be extracted successfully.

  The present invention has been invented in order to solve these problems in view of the above points, and an image processing apparatus that selects an area including an image element from a document image based on an intention of an operator, and An object is to provide an image processing method and a program.

  In order to achieve the above object, the image processing apparatus of the present invention employs the following configuration.

  The image processing apparatus of the present invention is an image processing apparatus that acquires a selection area from the processed image based on a specified area included in the processed image specified for the processed image, and indicates the range of the specified area Designated area information obtaining means for obtaining designated area information; first candidate selection area obtaining means for obtaining a first candidate selection area that is the largest area that can be the selected area based on the designated area information; A second candidate selection area acquiring unit that divides the candidate selection area for each type of image element included in one candidate selection area and acquires a second candidate selection area; and an image element included in the second candidate selection area. And a selection area acquisition means for acquiring the selection area by integrating the areas of the image elements based on the selection area.

  In the image processing apparatus, the first candidate selection region acquisition unit includes a rectangular region acquisition unit that acquires a rectangular region including the specified region based on the specified region information, and vertical and horizontal directions of the rectangular region. And a range search means for searching for a maximum area range that can be the selection area based on the presence or absence of pixels and determining the first candidate selection area.

  In the image processing apparatus, the second candidate selection region acquisition unit includes a block region dividing unit that divides the first candidate selection region into block regions, and a histogram that performs density histogram measurement for each of the divided block regions. Based on the measurement means, the measured density histogram, the type of the image element included for each block area, and the block area type determination means for determining the area type of the block, and the area type for each block And a homogeneous block acquisition means for acquiring a homogeneous block area consisting of the same area type.

  Further, in the image processing device, the second candidate selection area acquisition unit, when a plurality of similar block areas are acquired, based on the designated area, the same candidate block area from a plurality of the same block areas And determining means for determining as two candidate selection areas.

  In the image processing apparatus, the selection area acquisition unit includes an extraction unit that extracts an image element included in the second candidate selection region, and an integration unit that integrates a circumscribed region of the image element. And

  In order to solve the above-described problem, the present invention further provides an image processing method including steps for realizing the functions of the respective units included in the image processing apparatus, or a computer program for causing a computer to execute the image processing method. It is good.

  According to the image processing apparatus of the present invention, it is possible to provide an image processing apparatus, an image processing method, and a program for selecting a region including an image element from a document image based on an intention of an operator.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<Configuration example of image processing apparatus>
FIG. 4 is a configuration diagram of the image processing apparatus 1 according to an embodiment of the present invention and apparatuses connected to the image processing apparatus 1. 4 includes a CPU 101, a scanner unit 102, an external storage unit 103, an input image memory unit 104, a display unit 105, an operation panel unit 106, an area information storage unit 107, an image processing unit 108, and a communication control unit 109. The functions of the image processing apparatus 1 are realized through a bus.

  The CPU 101 performs overall control. The scanner unit 102 controls an apparatus that reads an image and acquires an image. The external storage unit 103 is connected to a drive of a storage medium such as the storage medium 111 and a storage device such as the hard disk drive 112, and these storage devices are controlled by the CPU 101. The image memory unit 104 stores image data processed by the image processing apparatus 1 including image data input from the scanner unit 102 or the like.

  The display unit 105 displays the status of the image processing apparatus 1 and further displays a preview of the processed image. The operation panel unit 106 is a device through which an operator inputs processing instructions and the like. The operation panel unit 106 also accepts designation of an area input by the operator. The display unit 105 and the operation panel unit 106 may be configured as one operation panel.

  The area information storage unit 107 is a memory that stores information (designated area information) related to designation of an area input from the operation panel unit 106 or the like. Although the designated area information will be described later, it is information such as coordinates representing a rectangle or a closed curve input by the operator.

  The image processing unit 108 performs processing for realizing the image processing method according to the embodiment of the present invention. Although details of this process will be described later, this process is a process of acquiring a selection area to be finally acquired by removing an excess part (area) or complementing an insufficient part based on designated area information. .

  The communication control unit 109 controls communication between the image processing apparatus 1 and other devices connected via the network 110. The communication control unit 109 inputs / outputs image data to / from another device or inputs an area designation from another device.

  The image processing apparatus 1 may be realized by, for example, a personal computer (PC), an MFP (Multi Function Peripheral), a facsimile (FAX), a digital camera, a mobile phone, a car navigation system, or the like.

<Function>
FIG. 5 is a functional block diagram showing main functions of the image processing apparatus 1 according to the embodiment of the present invention. The image processing apparatus 1 has functions according to the present invention as main functions, that is, a designated area information acquisition unit 501, a rectangular area acquisition unit 502, a range search unit 503, a block area division unit 504, a histogram measurement unit 505, and a block area. A type determination unit 506, a homogeneous block acquisition unit 507, an extraction unit 508, and an integration unit 509 are included. Each of these functional units can be said to be a function performed by the image processing unit 108 described above. It should be noted that each functional unit can be grouped from a functional aspect in terms of explanation. As the functional unit group, (designated region information acquisition unit 501), rectangular region acquisition unit 502, and range search unit 503 are used as functional unit groups for acquiring the first candidate selection region. The block region dividing unit 504, the histogram measuring unit 505, the block region type determining unit 506, and the same type block acquiring unit 507 are set as a functional unit group for acquiring the second candidate selection region. The extraction unit 508 and the integration unit 509 are function unit groups for acquiring a selection region to be finally output. Hereinafter, each functional unit will be described.

(Function unit group for acquiring the first candidate selection area)
The designated area information acquisition unit 501 has a function of obtaining designated area information indicating the range of the designated area. The designated area is an area designated by the user for the processed image and desired to be acquired. Specifically, the input designation area in FIGS. 2 and 3 corresponds to this. The designated area information is information indicating the designated area, for example, coordinate information corresponding to the designated area.

  The rectangular area acquisition unit 502 has a function of acquiring a rectangular area that includes and touches the specified area based on the specified area information acquired by the specified area information acquisition unit 501. For example, the input designation area in FIG. 3 is designated by a device such as a mouse or a touch pen, and is not rectangular. Therefore, the processing is performed on a rectangular area including the designated area. Specific details will be described later.

  Based on the rectangular area acquired by the rectangular area acquisition unit 502, the range search unit 503 selects the (first) candidate selection area that is the maximum area that can be a selection area (to be finally acquired). It has a function to decide. The candidate selection area refers to an area that can be provisionally selected as a selection area candidate to be finally acquired by image processing according to the present invention. For example, the input designation area in FIG. 3 is for obtaining a candidate selection area that is the maximum area that can be a selection area to complement the specified area because there is a portion that is insufficient in the designated area. Although specific details will be described later, the maximum area range that can be the selection area is searched based on the presence or absence of pixels in the vertical and horizontal directions of the rectangular area.

(Function unit group for acquiring the second candidate selection area)
The block area dividing unit 504 has a function of dividing the (first) candidate selection area acquired by the range searching unit 503 into block areas. This is because the image processing unit is handled in units of blocks.

  The histogram measuring unit 505 has a function of performing density histogram measurement for each block area divided by the block area dividing unit 504. This is for determining the area type (attribute) of the block area.

  The block area type determination unit 506 has a function of determining the type of image element included in each block area based on the density histogram measured by the histogram measurement unit 505 and determining the area type of the entire block. ing. The type of block area refers to what type of image (configuration) element the block area is, for example, an area mainly containing characters, an area mainly containing figures and photos, or an area mainly containing tables. The type of the area to be included.

  The same type block acquisition unit 507 has a function of acquiring the same type block area composed of the same area type based on the area type for each block area determined by the block area type determination unit 506. Thereby, for example, the area can be distinguished for each type, such as an area mainly including characters, an area mainly including drawings and photographs, and an area mainly including tables.

(Function group for acquiring selected area)
The extraction unit 508 has a function of extracting image elements included in the second candidate selection area. This is for performing the integration process.

  The integration unit 509 has a function of integrating image element regions (circumscribed regions). Although details will be described later, the purpose is to remove an unnecessary area that protrudes from a specified area. For example, this is for finally acquiring a selection region as shown in FIG.

  As described above, these functions are actually realized by a program executed by the CPU 101.

<Operation>
FIG. 6 is a flowchart for explaining an outline of the operation of the image forming apparatus 1 according to the present invention. As described above, for example, when the selection area is acquired, in the input designation area (FIG. 3) by the user, a protruding area, an extra area, or rather an insufficient area is generated. The image forming apparatus 1 according to the present invention performs an operation of cutting out only a necessary area (selected area) that does not include a useless area from an input designation area by a user in a document image in order to solve such a problem. is there. Hereinafter, description will be made along the flowchart with reference to the drawings as appropriate.

  In step S601, image data is input via a file, scan, or network (image input means). When applied to a web page or the like obtained from the Internet, a step of imaging it can be included.

  Further, the image data can be subjected to binarization processing (binarization means) in order to facilitate later-described processing. The binarization means is a means for binarizing the input processed image with a predetermined threshold. The binarization means may perform binarization processing based on the distribution of pixel values of the processed image, etc., or may perform binarization processing using a predetermined threshold value.

  In step S602, the designated area information acquisition unit 501 obtains designated area information. When the user designates an area that the user wants to take out with a mouse, a touch panel, or the like, or an area that is closed so as to surround the image components (designated area), the designated area in the image forming apparatus 1 is, for example, coordinate information (designated area information). And stored in the area information storage unit 107 described above. Each coordinate information (data) of the designated area can be expressed as S1 (x, y), S2 (x, y),... Sn (x, y). Note that the closed region includes shapes such as a circle, an ellipse, a closed curve, and a polygon.

  In step S <b> 603, the rectangular area acquisition unit 502 acquires a rectangular area that includes the specified area as a rectangle, i.e., covers the specified area. Specifically, among coordinate information of a plurality of designated areas, the x coordinate is the smallest value (a), the x coordinate is the largest value (b), the y coordinate is the smallest value (c), A value (d) having the largest coordinate is obtained, and rectangular vertex coordinate information is obtained based on the value. The coordinate information (a, c), (a, d), (b, c), (b, d), a, b, c, d> 0 thus obtained is a rectangle with four vertices (corners). In other words, this is a rectangular area including the specified area in contact with the specified area. This is shown in FIG. FIG. 7 is an example of a diagram illustrating a rectangular area. In addition, when the designated area information already indicates a rectangle, it itself is equal to the rectangular area.

  In step S604, the range search unit 503 determines a first candidate selection region that is the maximum region that can be a selection region (to be finally acquired) based on a rectangular region that includes the specified region in contact with the region. Note that “first” is a convenient expression for distinguishing from a “second” candidate selection region described later. Specifically, in order to determine a candidate selection area that is the maximum area that can be a selection area based on rectangular area coordinate data, a search range is determined in the vertical and horizontal directions in the coordinates. FIG. 8 is a diagram for explaining determination of the first candidate selection area. The search ranges Hwl (leftward), Hwr (rightward), Hht (upward), and Hhb (downward) in the vertical and horizontal directions can be determined by the following histogram measurement procedure.

(1) Obtain histograms of upper, lower, left and right pixels.
In the case of Hwl, the number of black pixels in the y direction from the x direction is obtained.
In the case of Hwr, the number of black pixels in the y direction with respect to the x direction is obtained.
In the case of Hht, the number of black pixels in the x direction from the y direction is obtained.
In the case of Hhb, the number of black pixels in the x direction from the y direction is obtained.

(2) It is determined whether or not the blank interval Sp (predetermined threshold) is exceeded.
When the number of black pixels is 0, the histogram measurement is aborted if the histogram of the coordinates before that coordinate is continuously 0 and exceeds the blank space Sp.

  FIG. 9 is a diagram illustrating how the search range is determined. That is, the result obtained by such a procedure is shown. As coordinate information, based on the coordinate information of the rectangular area, the coordinate information (a-Hwl, c-Hhb), (a-Hwl, d + Hht), (b + Hwr, c-Hhb) of the first candidate selection area ), (B + Hwl, d + Hht). In the figure, the search range Hwl is 0 because there is no black pixel.

  In step S605, the block area dividing unit 504 divides the first candidate selection area acquired by the range searching unit 503 into block areas. In the figure, the large outer frame line is the outer frame line of the first candidate selection region that is determined and acquired through a range search in the previous step. The first candidate selection area is divided into block areas at arbitrary intervals N in the horizontal direction. The arbitrary interval N is about 15 pt, but is not specified with this value. This is shown in FIG. FIG. 10 is an example of a diagram illustrating block area division.

  In step S606, the histogram measurement unit 505 obtains a density histogram of black pixels (or white pixels) for each divided block area. This is shown in FIG. FIG. 11 is an example of a diagram showing density histogram measurement for each block region. In the block area, an example of a density histogram of a character area and a graphic area is shown.

  In step S607, the block region type determination unit 506 determines the type of image element included in each block region based on the measured density histogram, and determines the region type of the block region. As described above, the area type of a block area refers to any type of image (for example, an area mainly including characters, an area mainly including drawings and photographs, an area mainly including tables). Configuration) This is the type of region that contains the element.

  Here, the density histogram has a unique tendency (feature) for each region type. In other words, the text area tends to have multiple valleys where the density histogram has a sufficient depth relative to the height of the peaks, and whether the density histogram is generally gentle in the figure / photo area. Alternatively, it can be said that the density histogram tends to have a curve close to a straight line in the table region. A specific example will be briefly described below.

  FIG. 12 is an example of a diagram illustrating a density histogram of a character image region. (A) is an example of a character image, (b) is an enlarged display of the “So” portion of the character image, and (c) is an image of the “So” portion displayed as a histogram of the number of black pixels in the y direction. is there. FIG. 13 is an example of a diagram showing a density histogram of a graphic (photo) image area. (A) is an example of a photographic image, (b) is an enlarged display of the eye portion of the photographic image, and (c) is a histogram of the number of black pixels in the y direction. FIG. 14 is an example of a diagram showing a density histogram of a table area. (A) is an example of an image of the front portion, (b) is an enlarged display of the line portion of the front portion, and (c) is a display of the number of black pixels in the y direction of the line portion as a histogram. It can be seen that there is a tendency as described above depending on the type of image element for each region.

  The block area type determination unit 506 uses these tendencies of the density histogram to make the block area an area mainly including characters (character area), an area mainly including figures and photographs (graphic area), and a table. A determination is made as to whether the block area is included (table area), and the area type of the block area is determined. FIG. 15 is an example of a diagram illustrating a state where the variation point of the density histogram is acquired. A graph obtained by drawing a line only at the changing point is acquired, and the determination is made based on the above-described tendency.

In the determination, further leveling processing can be performed. The density histogram has several fluctuating points, and there is a risk of misjudgment in some cases. This is because it is possible to easily determine the type of the region by leveling the density histogram by taking several fluctuation points as error ranges. Specifically, a plurality of variation points Dn are obtained from the density histogram for black pixels. Here, the change point is a change point where the element value decreases (increases) with a certain value as the element value that has been increased (decreased). The difference between the change point Dn and the next change point Dn + 1 is obtained, and the difference value obtained as an absolute value is defined as the change value En of the change point (D _n −D _{n + 1} = E ₁ ). Specifically,
D ₁ -D ₂ = E ₁
D ₂ -D ₃ = E ₂
D ₃ -D ₄ = E ₃ ...
It becomes. When the transition value En is lower than the specified value (threshold value) G, this transition value En is excluded. The designated value G is at least smaller than the interval N. In this way, the density histogram can be leveled. FIG. 16 is an example of a diagram illustrating an example of leveling of the density histogram. It is a graph in which the change points with small transition values are excluded and lines are drawn only at the change points with large transition values. From the number and distribution of the remaining transition values En, the block area can be accurately determined as a character area, a graphic area, or a table area.

  In step S608, the homogenous block acquisition unit 507 acquires the homogenous block area composed of the same area type based on the determined area type for each block area. FIG. 17 is an example of a diagram illustrating a state in which the blocks are acquired in the same type block area. Each block area is combined with the same type, and the character area and the graphic area are the same block area. The acquired same-type block area is set as a second candidate selection area.

  In the present embodiment thus far, two homogeneous block areas (character area and graphic area) have been acquired. When a plurality of identical block areas are acquired in this way, which homogeneous block area is retained. The problem is whether or not to keep both areas. That is, whether the character area is the second candidate selection area or the graphic area is the second candidate selection area. Here, in view of the input designation area in FIG. 3, the area intended by the user is a character area. This is because the character area occupies most of the designated area. Therefore, referring to each block area determined as a character area, a graphic area, and a table area, it is determined which type (attribute) is the most in the specified area, and the same type block composed only of that type (attribute) The region is finally determined as the second candidate selection region and acquired.

  Further, when it is desired to acquire both the character area and the graphic area as the user's intention, the proportion of the character area and the graphic area in the designated area is close by referring to the input designated area. In such a case, if a threshold is provided and the ratio is a predetermined occupancy ratio, it is determined that the user's intention includes both the character area and the graphic area, and the area obtained by combining the character area and the graphic area is finally the first. It can also be acquired as two candidate selection areas. The reason why the character area and the graphic area are combined is to allow the character area and the graphic area to be handled as a single unit in the integration process described later.

  In the present embodiment, the former, that is, the area intended by the user is a character area in view of the input designation area shown in FIG. 3, and therefore, a homogeneous block area having only the character area is finally acquired as the second candidate selection area. .

  As described above, the second candidate selection region is acquired by the processing so far. With the second candidate selection area, despite the rough area designation of the user as shown in FIG. 3, the “part lacking the desired area” is complemented by the outer area search, block area type determination, etc. It can be said that. However, the second candidate selection area is not “removed”. Accordingly, the selection area to be finally obtained is obtained by performing the process of “removing the excess part” by the subsequent process.

  Here, returning to step S605, a supplementary explanation will be given. In step S605, the first candidate selection area is divided into block areas at arbitrary intervals N in the horizontal direction. Assuming that characters and the like are written horizontally, it may be divided into block regions only in the horizontal direction as described above. Conversely, if characters and the like are written vertically, it is divided into block regions only in the vertical direction (for example, FIG. 18). Reference) The same processing as in the horizontal direction may be performed. Also, as a premise, if it is not clear whether the character direction is horizontal (horizontal writing) or vertical (vertical writing) in the document image, the above processing is performed on the block areas in both directions by dividing into block areas in both directions. Can do. That is, regardless of whether the character direction of the original image is horizontal writing or vertical writing, first, block areas are formed in both directions, and histogram measurement processing is performed. At the stage of the histogram measurement process, the horizontal histogram measurement process measures the distribution of pixels in the x direction and determines the x direction region from the result (this process does not look at the y direction). In the histogram measurement process in the vertical direction, the distribution of pixels in the y direction is measured, and the area in the y direction is determined from the result (this process does not look at the x direction). Then, the result of each direction may be integrated (synthesized) to obtain a similar block area as shown in FIG.

  Returning to FIG. 6 again, in step S609, the extraction unit 508 extracts image elements included in the second candidate selection region. It can also be said that a circumscribed area (or connected area) is extracted. This is because the integration process is performed for the purpose of removing unnecessary areas that protrude from the designated area. Specifically, for example, a plurality of circumscribed areas are extracted by a rectangular extraction method represented by the above-mentioned “Patent Document 1” (Japanese Patent Laid-Open No. 3-144873). FIG. 19 is a diagram illustrating a state in which an image element (circumscribed region) is extracted from the second candidate selection region. That is, a character or a circumscribed region of a character is extracted from a plurality of image elements included in the second candidate selection region. A circumscribed area (connected area) refers to an area where pixels are connected regardless of characters, lines, and figures.

  In step S610, the integration unit 509 integrates the circumscribed areas of the image elements, and acquires the integrated area. That is, in the previous step, selection is made so as not to exclude or exclude a plurality of circumscribed areas, and the remaining circumscribed areas are extracted, and integration processing (merging) of the circumscribed areas is performed. By this integration processing, “excess portions can be removed” from the second candidate selection region. Then, this integrated region obtained by removing the excess portion from the second candidate selection region can be a selection region to be finally acquired. FIG. 20 is a diagram illustrating an integrated area.

  As a supplement, since the present embodiment is configured to perform binarization processing of image data, accurately, an area corresponding to the integrated area acquired by the integrating unit 509 is acquired from the processed image (image data). To do. Of course, if the binarization process is not executed, this process may not be performed. In this case, the integrated area generated and acquired by the integrating unit 509 is the selection area to be finally acquired. . When the binarization process is not executed, the image processed by each of the above-described function units is an image based on the input image that has not been binarized, and thus the image output from the integration unit 509 is the selected region itself. That's why.

  In step S609, a circumscribed area (excluded range) that is not integrated in the next integration process can be considered according to the shape of the circumscribed area (for example, square or elongated). When integrating regions of image elements, for example, when not including elongated image elements, it may not be desired to include them. In the case where the character string is surrounded by a frame line so as to surround the character string, it may be desired to exclude a circumscribed area (connected area) indicating a long and narrow side line indicating the frame line. In other words, when the image element included in the designated area is a character, the integrated area may not include a frame line or a ruled line. Whether or not the area of the image element is elongated is determined using the formula shown in FIG. However, (width | height) is a value having a large value in the horizontal length or vertical length of the region of the image element, and n and α are parameters determined by an operator or the like. For example, it may be different based on the properties of the image. Of the horizontal length or vertical length of the area of the image element, when the length of the long side is larger than l1 and the length of the short side is smaller than l2, the image element is an elongated area. It is judged that. In this way, it is possible to set what kind of image element is to be extracted according to the usage status of the user.

  As described above, according to the image processing apparatus 1 according to the present invention, even if there is a deficient area or an extra area in the input designated area in the document image, the deficient area is complemented, and the extra area is Since it is removed, an appropriate rectangular area can be extracted from the document image as the selected area. That is, it is possible to provide an image processing apparatus, an image processing method, and a program for selecting a region including an image element from a document image based on the operator's intention.

  Although the best mode for carrying out the invention has been described above, the present invention is not limited to the embodiment described in the best mode. Modifications can be made without departing from the spirit of the present invention.

  For example, in the present embodiment, the case where it is desired to extract a character portion has been described. However, since not only the characters in the document image but also the graphic (illustration) and the table area are processed, for example, if the object related to the designated area is a graphic area, the same processing is performed on the graphic area. Yes. On the contrary, processing for excluding the character area is performed.

It is an example of the figure which shows a mode that a required area | region is cut out from the document image. It is an example of the figure which shows a mode that a required area | region is cut out from the document image. It is an example of the figure which shows a mode that a required area | region is cut out from the document image. 1 is a configuration diagram of an image processing apparatus 1 and an apparatus connected to the image processing apparatus 1 according to an embodiment of the present invention. It is a functional block diagram which shows the main functions of the image processing apparatus 1 which concerns on one Embodiment of this invention. 3 is a flowchart for explaining an outline of the operation of the image forming apparatus 1 according to the present invention. It is an example of the figure which shows a rectangular area. It is a figure explaining determination of the 1st candidate selection field. It is a figure which shows a mode that the search range was determined. It is an example of the figure which shows block area division | segmentation (horizontal). It is an example of the figure which shows density histogram measurement for every block area. It is an example of the figure which shows the density histogram of a character image area | region. It is an example of the figure which shows the density histogram of a figure (photograph) image area. It is an example of the figure which shows the density | concentration histogram of a table area. It is an example of the figure which shows a mode that the fluctuation | variation point of a density histogram is acquired. It is an example of the figure which shows an example of leveling of a density histogram. It is an example of the figure which shows a mode that it integrated and acquired by the same kind block area | region. It is an example of the figure which shows block area division | segmentation (vertical). It is a figure which shows a mode that an image element (circumscribed area | region) is extracted from a 2nd candidate selection area | region. It is a figure which shows an integrated area | region. It is a figure which shows an example of the determination formula whether the area | region of an image element is elongate.

Explanation of symbols

1 Image processing apparatus 101 CPU
102 Scanner unit 103 External storage unit 104 Image memory unit 105 Display unit 106 Operation panel unit 107 Area information storage unit 108 Image processing unit 109 Communication control unit 110 Network 111 Storage medium 112 Hard disk drive 501 Designated area information acquisition unit 502 Rectangular area acquisition unit 503 Range search unit 504 Block region division unit 505 Histogram measurement unit 506 Block region type determination unit 507 Same block acquisition unit 508 Extraction unit 509 Integration unit

Claims (11)

An image processing apparatus that acquires a selection area from the processed image based on a specified area of the processed image specified for the processed image,
Designated area information obtaining means for obtaining designated area information indicating a range of the designated area;
First candidate selection area acquisition means for acquiring a first candidate selection area, which is the maximum area that can be the selection area, based on the designated area information;
Second candidate selection area acquisition that divides the first candidate selection area for each type of image element included in the first candidate selection area, and acquires the same type area having the same image element type as the second candidate selection area Means,
Selection area acquisition means for integrating the areas of the image elements based on the image elements included in the second candidate selection area and acquiring the selection area;
Have
The second candidate selection area acquisition means, when a plurality of the same kind of areas are acquired, the same kind area having the same type of image element as the designated area is selected as the second candidate selection area from the plurality of the same kind of areas. To get,
An image processing apparatus. When there are a plurality of types of image elements included in the designated region, the second candidate selection region acquisition unit selects the same type region that is the same type as the type of the image element most frequently included in the designated region. Getting as an area,
The image processing apparatus according to claim 1. The first candidate selection area acquisition means includes
A rectangular area acquisition means for acquiring a rectangular area including the specified area based on the specified area information;
A range search means for searching for a maximum area range that can be the selection area in the vertical and horizontal directions of the rectangular area based on the presence or absence of pixels, and determining the first candidate selection area;
The image processing apparatus according to claim 1 or 2, characterized in that it has a. The second candidate selection area acquisition means includes
Block area dividing means for dividing the first candidate selection area into block areas;
A histogram measuring means for measuring a density histogram for each of the divided block areas;
Based on the measured density histogram, a type of image element included in each block region is determined, and a block region type determination unit that determines a region type of the block;
Based on the area type for each block, the same type block acquisition means for acquiring the same type block area consisting of the same area type,
The image processing apparatus according to any one of claims 1 to 3, characterized in that it has a. The selection area acquisition means includes
Extraction means for extracting image elements included in the second candidate selection area;
An integration means for integrating a circumscribed area of the image element;
The image processing apparatus according to any one of claims 1 to 4, characterized in that it has a. An image processing method in an image processing apparatus for acquiring a selection region from the processed image based on a specified region included in the processed image specified for the processed image,
A designated area information obtaining procedure for obtaining designated area information indicating a range of the designated area;
Computer
A first candidate selection area acquisition procedure for acquiring a first candidate selection area, which is the maximum area that can be the selection area, based on the designated area information;
Second candidate selection area acquisition that divides the first candidate selection area for each type of image element included in the first candidate selection area, and acquires the same type area having the same image element type as the second candidate selection area Procedure and
A selection area acquisition procedure for integrating the areas of the image elements based on image elements included in the second candidate selection area and acquiring the selection area;
Have
In the second candidate selection region acquisition procedure, when a plurality of the same type regions are acquired, the same type region having the same type of image element as the designated region is selected as the second candidate selection region from among the plurality of same type regions. To get,
An image processing method comprising: In the second candidate selection area acquisition procedure, when there are a plurality of types of image elements included in the designated area, the same type area that is the same type as the type of image element that is contained most in the designated area is selected as the second candidate selection area. Getting as an area,
The image processing method according to claim 6. The first candidate selection area acquisition procedure includes:
Based on the specified area information, a rectangular area acquisition procedure for acquiring a rectangular area including the specified area;
A range search procedure for searching the maximum area range that can be the selection area in the vertical and horizontal directions of the rectangular area based on the presence or absence of pixels, and determining the first candidate selection area;
The image processing method according to claim 6, further comprising : The second candidate selection area acquisition procedure includes:
A block region dividing procedure for dividing the first candidate selection region into block regions;
A histogram measurement procedures for the density histogram measurement for each of the divided said block area,
A block region type determination procedure for determining the type of image element included in each block region based on the measured density histogram and determining the region type of the block;
Based on the area type for each block, the same type block acquisition procedure for acquiring the same type block area consisting of the same area type,
The image processing method according to claim 6, further comprising : The selection area acquisition procedure includes:
An extraction procedure for extracting image elements included in the second candidate selection region;
An integration procedure for integrating the circumscribed area of the image element;
The image processing method according to any one of claims 6 to 9, characterized in that it has a. A program for causing a computer to execute the image processing method according to any one of claims 6 to 10 .

Images